Poster - PS1750EN00
Selection & Sizing of Clarification Depth Filters (IBC Biopharmaceutical Production Week, 2002)
Lit No:	PS1750EN00
Year:	2002

Scale-up of Prefilters

Important factors to consider for clarification of biological suspensions (e.g., mammalian or microbial cell cultures):

• Care in selecting filter
• Sizing method used in testing

Background

Three stages of cell culture clarification

• Primary: bulk removal of whole cells, cell fragments, and large particles.
• Typical methods:
• Tangential flow filtration microfiltration
• Centrifugation
• Secondary: removal of small particles (<1 µm), chiefly colloidal protein agglomerates.
• Typical method: Combination of depth and normal flow membrane filters
• Cellulosic, lenticular, stacked-disk depth filters are recognized as effective and cost efficient
• Sterile filtration: bioburden reduction and sterilization
• Typical method: Normal flow membrane filtration @ 0.22 micron

Mechanics of Depth Filtration

Two separation mechanisms

• Mechanical sieving (particle entrapment by smaller sized pores)
• Adsorption (particle binding due to electrostatic or other physio-chemical interactions with filter media)

• Sieving: pores or flow channels within the filter media will progressively clog with captured solids and the pressure differential across the filter element will gradually climb in response.
• “Pressure limited”- monitor prefilter pressure differential
• Adsorptive: no measurable change in pressure. The capacity or service life of the filter instead depends on the number of binding sites available. Once sites are saturated, the solids content of the filtrate will rise as the smaller feed particles pass through the filter unaffected.
• “Breakthrough limited” - monitor filtrate turbidity

• Monitoring both the prefilter pressure differential and filtrate turbidity will yield either a pressure limited or breakthrough limited performance profile, according to the dominant capture mechanism (see Figure 1).
• Test protocol that ignores the quality of the prefilter effluent by focusing exclusively on pressure build-up can give highly misleading results.
• Conduct the performance test with prefilter and trailing sterile filter operated in tandem allowing an engineer to critically evaluate the effects of operating conditions relative to filter sizing.

Performance Profile

Figure 1. Pressure-limited and breakthrough-limited performance profile.

Prefilter Scale-up

• Application: secondary clarification of cell culture suspensions (mammalian or microbial)
• Important factors to consider:
• Selection of prefilter type and materials
• Test method used in system sizing

Depth Filter Sizing

Depth filter Capacity

• Defined as the volume of fluid that can be processed up to a maximum allowable pressure differential (liters per square meter of filter frontal area).

• Volumetric flow rate per unit cross-sectional area (fluid velocity or flux rate) will have a significant negative effect on prefilter capacity the higher the flow rate, the lower the capacity.
• A constant-pressure test that generates a variable flow rate over time will typically result in an underestimation of filter capacity.
• Initial surge in feed flow can load the depth filter inefficiently and thereby create premature clogging and poor capacity

The Vmaxsm Method

• Constant pressure test: filtrate volume measured as a function of time.
• Features:
• Bench-scale, short duration test
• Predicts maximum filtrate volume attainable per unit filter area for estimating scale-up system sizing
• Based on gradual pore-plugging model
• Limitations of Vmax Method
• Inappropriate for depth filter media that do not comply with gradual pore-plugging model.
• Variable flow rate does not accurately assess depth filter capacity.
• Monitoring filtrate quality (turbidity) difficult

Effect of Flux Rate on Prefilter Capacity

Figure 2: Pressure Drop vs. Prefilter Capacity Effect of Flux Rate

Filter Sizing Example

Test Method

• A model feed solution of lyophilized dairy whey was filtered at a constant feed rate (Pmaxsm method) using a cellulosic depth filter (Millistak+® DE75). The differential pressure was monitored as a function of throughput or capacity. The test was repeated at four feed rates to note the effect on depth filter capacity.

Results

• As shown in Figure 2, lowering the feed rate (measured in liters per square meter per hour, LMH) substantially increased the capacity of the filter by slowing the rise in pressure. By comparison, a constant-pressure (Vmax method) test performed at 20 psid with the same feed material and prefilter gave a less-than-optimal capacity approximately, 40% less than the lowest flux rate tested.
• It is important to note the linear relationship between flow rate and capacity as demonstrated in the above tests. By normalizing the pressure measurements to the flow rate for any given test, the previous data are replotted in Figure 3 to depict prefilter resistance (psid/LMH) as a function of capacity.
• The result of the normalization merges the individual controlled-flow tests into a single performance curve. Such information can be used to estimate the capacity of a prefilter at various process flow rates and pressures of interest.

Conclusion

• The Vmax test fails to reflect the fact that higher capacities could be achieved at lower flow rates. Therefore, to get an accurate profile of a given depth filter, the Pmax method should be applied at several selected flow rates over a range that includes the lowest practical rate.

Resistance vs. Prefilter Capacity

Figure 3: Resistance vs. Prefilter Capacity

Process Scale-up

To verify the usefulness of the Pmax method, a scale-up study was undertaken on a typical recombinant protein (MAb) production process.

Prefilter + Sterile Filter Scale-up

Figure 4: Prefilter + Sterile Filter Scale-up

Sieving vs. Adsorption

Conclusion

• A constant-flow (rather than constant-pressure or Vmax) test method has been shown to offer a more accurate performance profile of depth filters used in the clarification of biological suspensions .
• Selection of the right depth filter according to media type and construction is equally important.
• Millistak+ HC demonstrates high capacity and retention for the wide variety of suspended particles and colloids that are present in fermentation broth. These unique stacked-disk depth prefilters owe their superior performance to a multilayer construction tailored to the specific demands of this challenging application.